Carotid Body Function in Tyrosine Hydroxylase Conditional Olfr78 Knockout Mice

Abstract The Olfr78 gene encodes a G-protein-coupled olfactory receptor that is expressed in several ectopic sites. Olfr78 is one of the most abundant mRNA species in carotid body (CB) glomus cells. These cells are the prototypical oxygen (O2) sensitive arterial chemoreceptors, which, in response to lowered O2 tension (hypoxia), activate the respiratory centers to induce hyperventilation. It has been proposed that Olfr78 is a lactate receptor and that glomus cell activation by the increase in blood lactate mediates the hypoxic ventilatory response (HVR). However, this proposal has been challenged by several groups showing that Olfr78 is not a physiologically relevant lactate receptor and that the O2-based regulation of breathing is not affected in constitutive Olfr78 knockout mice. In another study, constitutive Olfr78 knockout mice were reported to have altered systemic and CB responses to mild hypoxia. To further characterize the functional role of Olfr78 in CB glomus cells, we here generated a conditional Olfr78 knockout mouse strain and then restricted the knockout to glomus cells and other catecholaminergic cells by crossing with a tyrosine hydroxylase-specific Cre driver strain (TH-Olfr78 KO mice). We find that TH-Olfr78 KO mice have a normal HVR. Interestingly, glomus cells of TH-Olfr78 KO mice exhibit molecular and electrophysiological alterations as well as a reduced dopamine content in secretory vesicles and neurosecretory activity. These functional characteristics resemble those of CB neuroblasts in wild-type mice. We suggest that, although Olfr78 is not essential for CB O2 sensing, activation of Olfr78-dependent pathways is required for maturation of glomus cells.


Introduction
Olfactory receptor 78 (Olfr78) is a G-protein-coupled odorant receptor that is atypically expressed outside the olfactory epithelium.Olfr78 is expressed at high levels in the carotid body (CB), [1][2][3] the prototypical acute oxygen (O 2 ) sensor, and main organ responsible for the hypoxic ventilatory response (HVR). 4,5ne group reported that Olfr78 acts as a lactate receptor in mice and that the HVR in constitutive Olfr78 knockout mice (Olfr78 −/− mice) is abolished and the CB responses to hypoxia blunted. 1 Based on their findings, this group proposed a model whereby an increase in lactatemia during hypoxia and the subsequent endocrine activation of Olfr78-expressing CB glomus cells are responsible for the O 2 regulation of breathing. 1This model has been challenged by several groups showing that lactate is a poor agonist for Olfr78, 2,6 that Olfr78 is not a physiologically relevant lactate receptor, 7,8 and that the HVR is not significantly altered in several strains of Olfr78 −/− mice. 7It is well established that O 2 sensing is an intrinsic property of CB glomus (or type I) cells [9][10][11] and that these polymodal chemoreceptors are activated by low O 2 tension and other stimuli, including lactate, via mechanisms that are unrelated to Olfr78. 4,12Nonetheless, one study reported that Olfr78 −/− mice have impaired HVR and glomus cell responses to mild hypoxia 8 and a reduction in sympathetic activation and systemic hypertension induced by chronic intermittent hypoxia. 13lfr78 is one of the most abundant mRNA species expressed in CB glomus cells, but its functional significance remains poorly understood.As is the case for other genes relevant for CB glomus cell function, 14 Olfr78 mRNA levels also depend on the constitutively high expression of hypoxia-inducible factor 2 alpha (HIF2α). 15Therefore, it is conceivable that, although not essential for O 2 sensing, Olfr78 has a role in glomus cell homeostasis.Importantly, all previous reports were about mice with a constitutive (or global) knockout of the Olfr78 gene 16 and therefore the phenotypes observed could result from pleiotropic effects on the various CB cell types and/or on various organs. 17In addition, the properties of Olfr78-deficient CB glomus cells have not been analyzed in detail.
Here, we report the generation and characterization of a C57BL/6-inbred conditional Olfr78 knockout mouse strain, which we crossed with a gene-targeted TH-IRES-Cre driver strain in order to restrict the knockout of the gene to CB glomus cells and other catecholaminergic cell types.Offspring of this cross are henceforward referred to as TH-Olfr78 KO mice.We did not observe significant changes in the HVR of TH-Olfr78 KO mice.However, they exhibited a subtle but clearly reproducible glomus cell phenotype, with changes in mRNA expression, electrophysiological features, intracellular Ca 2+ dynamics, and secretory activity that suggest alterations in the trophic maintenance of these cells.We speculate that Olfr78 contributes to maintaining a high basal activity of the adenyl cyclase-cyclic AMP pathway, which may be required for phenotypic specification of mature glomus cells.

Ethical Approval
In Frankfurt, mouse experiments were carried out in accordance with guidelines of the German Animal Welfare Act, the Directive 2010/63/EU of the European Parliament and of the Council, and the institutional ethical and animal welfare guidelines of the Max Planck Research Unit for Neurogenetics.Approval came from the Regierungspra¨sidium Darmstadt (Germany), and the Vet-erina¨ramt of the City of Frankfurt (Germany).In Seville, the Institutional Committee for Animal Care and Use at the University of Seville (21/04/2020/053 and 27/01/2017/018) approved all procedures used in this study.Handling of the animals was conducted in accordance with the European Community Council directive of September 22, 2010 (Directive 2010/63/EU) and the implementations of June 5, 2019 (Regulation 2019/2010) for the care and use of laboratory animals.To minimize the number of animals used in the experiments, mice subjected to in vivo analyses were subsequently euthanized and used for the in vitro studies.As all measurements of physiological constants done on mice were painless, analgesia was not used.For in vitro experiments, mice were killed by intraperitoneal injection with a lethal dose of anaesthetic (sodium thiopental, 120-200 mg/kg)

Generation of Tyrosine Hydroxylase Conditional Olfr78 Knockout Mice
The coding region of exon 4 of the Olfr78 gene was flanked by loxP sites by gene targeting in the C57BL/6-derived embryonic stem cell line Bruce4.After obtaining germline transmission, the neo-selectable marker flanked by FRT sites was removed from the targeted mutation by crossing with Oz flp, a C57BL/6inbred strain that carries a knockin of the site-specific recombinase flp in the ROSA26 locus driven by the UbiC promoter; the Oz flp allele was crossed out subsequently.The strain carrying the floxed Olfr78 allele without the Oz flip allele is publicly available from The Jackson Laboratory as stock #32801 with official strain name B6.Cg-Or51e2<tm3.1Mom>.Mice carrying the conditional Olfr78 knockout allele in the heterozygous state (Olfr78 f/+) or homozygous state (Olfr78 f/f) were crossed with TH-IRES-Cre mice. 18In this Cre driver strain, the gene-targeted Th-IRES-Cre knockin mutation provides Cre expression without disturbing expression of the tyrosine hydroxylase (Th) gene and affords efficient in vivo Cre-mediated recombination in catecholaminergic cells. 19,20

Animal Care and Exposure of Mice to Chronic Hypoxia
Mice were housed at 22 • C ± 1 • C in a 12-h light/12-h dark cycle with ad libitum access to food (Teklad global 14% protein, Envigo) and water.Both male and female mice were used.In TH-Olfr78 KO mice maintained under chronic hypoxia (10% O 2 atmosphere for 14 d), the experiments were performed using a hermetic chamber with control of O 2 , CO 2 , humidity, and temperature (Coy Laboratory Products).For the experiments, mice (2-3 moold) were maintained in standard rodent cages placed into the hypoxia chamber, with ad libitum access to pellet food and water, and within a 12-h light/12-h dark cycle.

Plethysmography
Plethysmography was used to study respiratory function in conscious unrestricted mice as previously described. 23Briefly, mice were placed in plethysmographic chambers (EMKA Technologies) and perfused at a constant flow rate (1 L/min) with air (21% O 2 , normoxia), and various gas mixtures: 10% O 2 , 12% O 2 (hypoxia, maintained for 5 min once O 2 percentage reached 10% or 12%), or 5% CO 2 (hypercapnia, maintained during 1 min when CO 2 percentage reached 5%).The hermetic chambers were provided with O 2 and CO 2 sensors to monitor the gas composition in parallel with changes in respiratory frequency recorded by a pressure sensor during the experiment.Data acquisition was performed using Iox2 (RRID:SCR 022973; EMKA Technologies).To calculate changes in respiratory frequency, the basal, hypoxic, and hypercapnic respiratory frequencies were estimated in each animal.Basal respiratory frequency was calculated by averaging the values acquired during the 160 s previous to exposure to hypoxia.Peak respiratory frequency (indicated by "p" in the figures) during hypoxia was calculated by averaging the values obtained during 110 s at the peak of the hypoxic response.The total average respiratory frequency (indicated by "a" in the figures) during exposure to hypoxia (10% or 12% O 2 ) was estimated by the mean of values measured during the last 400 s of exposure to stimuli.Respiratory frequency during exposure to hypercapnia was estimated by averaging the values obtained during 90 s after reaching 4%-5% CO 2 in the chamber before returning to normoxia.

Hematocrit
Hematocrit was measured in blood samples collected from the carotid artery using a hematocrit tube located near the incision site (where the bifurcation was dissected) to allow blood to fill the tube by capillary action.Following blood collection, tubes were sealed with bone wax, placed in a microhematocrit centrifuge, and spun for 5 min at 3 g .Hematocrit was calculated manually by measuring the length of the column of packed red cells and total blood length and expressed as the percentage of erythrocytes relative to total blood volume (100%).

Immunohistochemistry and Estimation of CB Volume
For immunohistochemical studies, mice were first perfused with phosphate buffered saline (PBS) and then with 4% paraformaldehyde in PBS before tissue dissection.Carotid bifurcations were fixed with 4% paraformaldehyde in PBS for 2 h, cryoprotected overnight with 30% sucrose in PBS, and embedded in O.C.T. (Cat# 4583, Tissue-Tek).Tissue sections of 10 μm were cut with a cryostat (Leica, Wetzlar) and incubated overnight at 4 • C with primary antibodies: TH (1:100 dilution, Cat#AB1542, Sigma-Aldrich) and Ki67 (1:100 dilution, Cat#RM-9106, Epredia).This was followed by incubation with fluorescent secondary antibodies: Alexa Fluor 488 or Alexa Fluor 568 (1:500 or 1:1000 dilution, Cat#A11008 and Cat#A11004, Invitrogen, RRID: AB 143165 and RRID: AB 2534072).Nuclei were labeled with 4 ,6diamidino-2-phenylindole (DAPI).Immunofluorescence images were obtained using a Nikon A1R+ confocal microscope.To estimate CB volume, serial sections of CBs were immunostained and the area within each CB section was calculated using Image J software (RRID:SCR 003070).Carotid body volume was calculated considering CB area in each section, section thickness, and total number of sections per CB.

Carotid Body and Superior Cervical Ganglion Resection and Molecular Analyses
Mice were sacrificed by intraperitoneal administration of a lethal dose of sodium thiopental (120-150 mg/kg).Carotid bifurcations were removed and placed in cold PBS.Carotid bodies and superior cervical ganglion (SCG) were dissected.For real-time quantitative polymerase chain reaction (PCR) analysis, total RNA was isolated from CB and SCG using RNeasy Micro kit (Qiagen).For  the analysis of CB gene expression profile in TH-Olfr78 KO and wild-type (WT) mice, each CB replicate was pooled from 3 mice to obtain enough amount of mRNA.Complementary RNA (cRNA) was then amplified from CB total RNA using the GeneChip WT PLUS Reagent Kit (Affymetrix).Total RNA (500 ng; or amplified cRNA in the case of CB) was reverse-transcribed into complementary DNA (cDNA) using the QuantiTect Reverse Transcription Kit (Cat#205311, Qiagen).Real-time quantitative PCR was performed on a 7500 Fast Real Time PCR System (Applied Biosystem).PCR was performed in duplicate in a total volume of 20 μL containing 1 μL of cDNA solution and 1 μL of TaqMan probe of the specific genes (Thermo Fisher Scientific).Gapdh mRNA was measured in each sample to normalize the amount of total RNA (or cRNA) input to perform relative quantifications.
To prepare CB slices, dissected CBs were placed in ice-cooled modified Tyrode solution and processed as described previously. 25,26Briefly, CB were included at 42 o C in 1% low melting point agarose (prepared in PBS).Agarose block containing CBs were glued to the platform a vibratome (VT1000S, Leica) to prepare sections of 150 μm.Carotid body sections were digested in enzymatic solution (the same used for dispersed glomus cells), during 5 min, in a water bath at 37 o C with shaking.Finally, slices were washed with PBS and incubated in the same culture medium used for dispersed glomus cells to which 1.2 U/mL of erythropoietin (Cat#EU 1/07/410/028, Sandoz) were added.Slices were incubated at 37 • C in 5% CO 2 for 24 h before use.

Amperometry
To monitor single-cell secretory activity, dopamine secretion from glomus cells in CB slices was measured by amperometry. 23ecretory events were recorded with a 10 μm carbon fiber electrode polarized to +750 mV (to favor dopamine oxidation) using  an external voltameter connected to the EPC-7 amplifier.Amperometric currents were recorded with an EPC-7 patch-clamp amplifier (HEKA Electronics, Lambrecht/Pfaltz).The signal was filtered at 100 Hz and digitized at 250 Hz before storage on computer.Data acquisition and analysis were carried out with an ITC-16 interface (Instrutech Corporation) and PULSE/PULSEFIT software (HEKA Electronics).For the experiments, a slice was transferred to the recording chamber of an upright microscope (Axioscope, Zeiss) and continuously perfused with extracellular solution (see the "Recording Solutions" section).The secretion rate (given in fC/min or pC/min) was calculated as the amount of charge transferred to the recording electrode during a given period of time.Experiments were performed at ∼35 • C

Microfluorimetric Measurements of Intracellular Ca 2+
Microfluorimetric measurement of intracellular changes in Ca 2+ concentration was performed in single dispersed glomus cells as previously described. 11,19Dispersed CB cells were loaded with Fura2-AM (TefLabsMW1002, Molecular Probes), 4 mm in DMEM/F-12 without serum at 37 • C for 30 min and subsequently incubated for 15 min in complete medium to remove excess Fura2-AM.To perform the experiments, a coverslip with Fura 2-AM loaded cells was placed on a recording chamber mounted on the stage of a microscope equipped with epifluorescence and photometry.This set up consists of an inverted microscope (Nikon Eclipse Ti) equipped with a 40x/0.60NA objective, a monochromator (Polychrome V, Till Photonics), and a CCD camera, controlled by Aquacosmos software (version 2.6, Hamamatsu Photonics).Alternating excitation wavelengths of 340 and 380 nm with emission wavelength of 510 nm were used to obtain the F340/F380 ratio. 11Background fluorescence was subtracted before obtaining the F340/F380 ratio.A dichroic FF409-Di03 (Semrock) and a band-pass filter FF01-510/84 (Semrock) were used.Cytosolic Ca 2+ signals were digitized at a sampling interval of 500 ms.Experiments were performed at ∼35 • C.

Patch Clamp Recording and Electrophysiological Analyses
Macroscopic ionic currents were recorded from dispersed mouse glomus cells using the whole cell configuration of the patch clamp technique as adapted in our laboratory. 24,27Patch clamp electrodes (2-3 M ) were pulled from capillary glass tubes (Kimax, Kimble Products) with a horizontal pipette puller (Sutter instruments model P-1000) and fire-polished with a microforge (MF-830, Narishige).Voltage-clamp recordings were obtained with an EPC-7 patch clamp amplifier (HEKA Electronik) using standard voltage-clamp protocols designed with PULSE/PULSEFIT software (HEKA Electronik).The signal was filtered (10 kHz), subsequently digitized with an analog/digital converter (ITC-16 Instrutech Corporation), and finally sent to a computer.Data acquisition and storage were performed using the PULSE/PULSEFIT software (HEKA, Electronics) at a sampling interval of 20 μs.Experiments were performed at ∼35 • C.

Recording Solutions
Dispersed cells or slices used for in vitro amperometric or microfluorimetric recordings were transferred to the recording chamber and continuously perfused with a control solution containing 125 mm NaCl, 4. To monitor macroscopic Ca 2+ , Na + , and K + currents, glomus cells were perfused with external solutions and dialyzed with internal solutions.Solutions used to record whole-cell currents through Na + and Ca 2+ channels (using Ba 2+ as a charge carrier) contained the following: external solution: 140 mm NaCl,10 mm BaCl 2 , 4.7 mm KCl, 10 mm Hepes, and 10 mm glucose (pH 7.4) (osmolality, 300 mOsm/kg); internal solution: 130 mm CsCl, 10 mm EGTA, 10 mm Hepes, and 4 mm adenosine triphosphate (ATP-Mg) (pH 7.2; osmolality, 285 mOsm/kg).In some experiments, a 0 Na + solution was used, in which external Na + was completely replaced with the impermeant cation N-methyl Dglucamine.The external solution used to record whole-cell K + currents was the same control solution used for amperometric and microflourimetric analysis, and the internal solution contained 80 mm K + glutamate, 50 mm KCl, 1 mm MgCl 2 , 10 mm Hepes, 4 mm ATP-Mg, and 5 mm EGTA (pH 7.2).In the high K + solutions, KCl replaced NaCl equimolarly.In experiments where short chain lipid compounds were used, they also replaced NaCl equimolarly at the concentrations indicated.

Statistical Analysis
Statistical analysis was carried out using Prism Version 8.

Expression of Genes Relevant to CB Function in TH-Olfr78 Knockout Mice
Mice carrying the newly generated floxed Olfr78 allele were crossed with gene-targeted knockin mice expressing Cre recombinase under the control of the tyrosine hydroxylase (Th) promoter. 18,19The resulting mice (TH-Olfr78 KO mice; Figure 1A) showed a complete absence of Olfr78 mRNA in CB glomus cells (Figure 1B) and other catecholaminergic tissues such as the SCG (Figure 1C).Next, we studied differences in the mRNA expression levels of a set of genes that are known to be relevant for CB glomus cell function. 2,3,14,28,29These included Epas1 (encoding HIF2α), 3 atypical mitochondrial complex IV (MCIV) subunit isoforms (Cox4i2, Cox8b, and Higd1C), pyruvate carboxylase (Pcx), regulator of G-protein-signaling 5 (Rgs5), and tyrosine hydroxylase (Th), which is expressed in the highly dopaminergic mature glomus cells.Among these genes, Higd1C, which has been suggested to contribute to the characteristically high sensitivity of glomus cell mitochondria to hypoxia, 28 was significantly upregulated in Olfr78-deficient cells (Figure 1D and E).There was also a trend for Th mRNA to be downregulated in TH-Olfr78 KO mice but the data do not reach statistical significance (Figure 1E).As is the case for other genes critically involved in adult CB function, 14,15 Olfr78 mRNA expression appears to be dependent on HIF2α (Figure 1F).

In Vivo Responses to Acute and Chronic Hypoxia
The CB is the prototypical acute O 2 -sensing organ in adult mammals and main organ that is responsible for the HVR. 4,5To evaluate the responsiveness of WT mice and TH-Olfr78 KO mice to hypoxia, we tested the acute effects of O 2 /CO 2 on ventilation and the changes in CB structure induced by sustained hypoxia.
Mice exposed acutely to either 10% or 12% O 2 responded with rapid graded increases in respiration rate that have a peak (lasting for ∼1 min) at the onset of the response, followed by a progressive decline due to the inhibitory effect of hypocapnia caused by hyperventilation (Figure 2A-D).Neither the peak ("p" in the figure) nor the average ("a" in the figure) increases in breathing frequency induced by lowering environmental O 2 tension from normal air (21% O 2 ) to either 10% O 2 or 12% O 2 are different between WT mice and TH-Olfr78 KO mice (Figure 2A-D).Exposure to hypercapnia (5% CO 2 in a 21% O 2 atmosphere) produced a robust and sustained increase in breathing frequency, which is also similar in the 2 mouse models (Figure 2E and F).These data confirm our findings in mice with a constitutive (or  global) knockout of the Olfr78 gene 7 and indicate that Olfr78 deficiency does not alter O 2 /CO 2 regulation of breathing within the range of O 2 tensions that are normally used to monitor the HVR.
The CB grows severalfold in size upon exposure to chronic (days/weeks) hypoxia. 30This well-known adaptive response ensures the sustained activation of the respiratory center that is required for acclimatization to hypoxic environments.Carotid body growth upon chronic hypoxia requires the activation of CB glomus cells, 31 which release transmitters that rapidly induce proliferation and maturation of neighboring TH-positive CB neuroblasts (immature glomus cells) followed by activation of pluripotent CB stem cells. 17,32We found that the number of THpositive cells, proliferating cells (Ki67 positive), and the CB volume are comparable in WT mice and TH-Olfr78 KO mice maintained in a hypoxic environment (10% O 2 ) for 2 wk (Figure 3A-D).Moreover, the increase in hematocrit after exposure to chronic hypoxia, a parameter that is augmented in mice with altered CB O 2 sensing and no hyperventilation in response to hypoxia, 20 is similar in WT mice and TH-Olfr78 KO mice (Figure 3E).Taken together, we did not detect any differences of note in the in vivo adaptive responses to sustained hypoxia between WT mice and TH-Olfr78 KO mice.

Electrophysiology of Glomus Cells
The electrical properties of dispersed glomus cells of WT mice and TH-Olfr78 KO mice were recorded with the whole-cell configuration of the patch clamp technique.Average passive electrical parameters (cell capacitance and input resistance) are similar in the two cell types, although there is a trend for cell capacitance (proportional to cell size) to be smaller in Olfr78deficient cells (Figure 4A and B).All cells showed robust macroscopic voltage-dependent K + currents.The time course of the K + currents varied among glomus cells of WT and TH-Olfr78 KO mice, suggesting that different subtypes of voltage-gated K + channels with variable activation and inactivation kinetics are expressed in these cells (Figure 4C).The peak K + current amplitude (presented as current density) at membrane potentials >10 mV tends to be larger in Olfr78-deficient cells (Figure 4D and E).Voltage-dependent inward Na + and Ca 2+ currents were recorded in WT mice and TH-Olfr78 KO mice glomus cells dialyzed with Cs + (to block K + channels).Application of 10-ms depolarization pulses induces in all cells inward Ca 2+ currents, which, in some cases, are preceded by a fast and transient current (Figure 5A-C).This transient component of the current is selectively abolished after replacement of external Na + with the impermeant cation N-methyl glucamine (I Na , Figure 5A), suggesting that it is due to the opening of rapidly inactivating Na + channels.The Ca 2+ current, which is unaffected by Na + removal, remains stable during the pulse (I Ca , Figure 5A).The number of cells expressing a measurable Na + current is larger in TH-Olfr78 KO mice (82%, n = 22 cells) than in WT mice (54%, n = 24 cells).
In addition, Olfr78-deficient cells have larger peak Na + current density (I Na at +10 mV) in comparison to WT cells (Figure 5D and  E).Small differences in the density of Ca 2+ channel currents (I Ca at +10 mV) between Olfr78-deficient cells and WT cells are not statistically significant (Figure 5F and G).The time constant and  amplitude of the fast and slow deactivating components of the tail currents (Figure 6A), representing the two main Ca 2+ channel classes expressed in glomus cells, [33][34][35] are not affected by Olfr78 deficiency (Figure 6B-E).Taken together, these data reveal clear differences in the level of Na + channel expression between CB glomus cells of WT and TH-Olfr78 KO mice.Olfr78-deficient cells appeared to also display subtle differences in capacitance (smaller cell size) and in K + and Ca 2+ current density, but the results are not statistically significant.

Neurosecretory Responses of Glomus Cells to Hypoxia and Other Stimuli
Glomus cells are O 2 -sensitive neurosecretory elements that respond to hypoxia with membrane depolarization, Ca 2+ influx, and an increase in cytosolic [Ca 2+ ].This Ca 2+ signal triggers the exocytotic release of transmitters that activate afferent sensory fibers impinging on brain respiratory centers. 30We monitored changes in cytosolic [Ca 2+ ] by microfluorimetry in Fura-2-loaded dispersed glomus cells.Brief exposure to an externally applied depolarizing solution with high K + induced smooth Ca 2+ signals of large amplitude, which developed in parallel with the time course of bath solution exchange.The amplitude of these signals is similar in glomus cells of WT mice and TH-Olfr78 KO mice (Figure 7A and B).We also observed, both in glomus cells of WT mice and TH-Olfr78 KO mice, the presence of highly reversible hypoxia-induced Ca 2+ signals that varied in amplitude and duration depending on the O 2 tension in the bath solution (hypoxia, ∼15 mmHg; 3% O 2 , ∼30 mmHg) (see representative traces in Figure 7C and D).In some cases, the hypoxiainduced Ca 2+ signal had an oscillatory pattern, which was more frequent in Olfr78-deficient glomus cells.Clear repetitive Ca 2+ spiking signals were observed in more than 70% of cells of TH-Olfr78 KO mice but only in ∼40% of cells of WT mice (Figure 7E).Quantification of the area under the Ca 2+ signals induced by hypoxia indicates that cells of TH-Olfr78 KO mice are less responsive to hypoxia than cells of WT mice (Figure 7F).
To further examine the neurosecretory responses of glomus cells in CB slices, we used amperometry to directly monitor the exocytotic release of dopamine (Figure 8A).This noninvasive assay allows to perform a quantitative evaluation of the intrinsic O 2 sensitivity of individual glomus cells. 27,30Both WT and Olfr78-deficient cells responded with graded secretory responses to variable levels of hypoxia in the bath solution (hypoxia: ∼15 mmHg; 3% O 2 : ∼30 mmHg; 6% O 2 : ∼50 mmHg; control 21% O 2 : ∼150 mmHg) (Figure 8B and C).However, the secretory responses of Olfr78-deficient cells are smaller than those of WT cells across the range of O 2 tensions tested; with differences statistically significant only for the lowest O 2 tensions (Figure 8D-G).As the secretion rate depends not only on the exocytotic activity (frequency of events) but also on the charge carried by each secretory event, we quantified the quantal content of individual events in glomus cells that were exposed only to a low level of activation (O 2 tension ∼50 mmHg) in order to prevent the fusion of vesicles in the cytosol before secretion (compound exocytosis) (Figure 8H).For context, we previously showed that glomus cells of HIF2α-deficient mice have a lower dopamine content per vesicle than WT cells. 14We found that Olfr78-deficient glomus cells also have a smaller mean charge per vesicle and a distribution of dopamine content in secretory granules displaced to smaller values compared to WT cells (Figure 8H and I).In addition to hypoxia, we tested the secretory response of these cells to challenges with high K + , hypercapnia and hypoglycemia, which are well-known glomus cell stimuli.Inappreciable differences between WT and Olfr78-deficient cells were seen for responses to high K + and hypercapnia (20% CO 2 ) (Figures 8B and 9A-C).However, there is a trend for Olfr78deficient cells to show a lower response to hypoglycemia than WT cells (Figure 9D and E).Together, these data show that the absence of Olfr78 affects the neurosecretory function of CB glomus cells.Olfr78-deficient glomus cells have altered cytosolic Ca 2+ dynamics, a reduction in dopamine content of secretory vesicles, and decreased secretory activity in response to hypoxia or hypoglycemia.7][38] Olfr78-deficient CB cells with robust responses to hypoxia and hypercapnia were also strongly activated by acetate or β-hydroxybutyrate, which are circulating metabolically relevant short-chain lipids (Figure 10A and B).Secretory activity induced by acetate or β-hydroxybutyrate was quantitatively similar in glomus cells of WT and TH-Olfr78 KO mice (Figure 10C).

Discussion
The first conclusion of this study is that, as we showed previously, 7 Olfr78 is not essential for the O 2 regulation of breathing.Olfr78 is not a physiologically relevant lactate receptor 7,8 and, despite its high level of expression in the CB, it is dispensable for both the HVR and CB growth during acclimatization to sustained hypoxia.Under our experimental conditions, we were not able to confirm the decrease in the ventilatory response to mild hypoxia (12% O 2 ) reported in global Olfr78 knockout mice. 8However, at the single glomus cell level, our results support previous observations indicating that CB sensory nerve firing is activated by short-chain lipid compounds, such as acetate or butyrate 8,39 and that, as occurs with lactate, 7,12 these responses are not affected in Olfr78 knockout mice. 8he second main conclusion of this study is that CB glomus cells of TH-Olfr78 KO mice exhibit in vitro complex phenotypic changes compared with WT cells.Although subtle in some respects, these are clearly identifiable.These changes included a trend for decreased Th expression, an increased expression of voltage-gated Na + channels and altered neurosecretory responses consisting of a decrease in secretory vesicle dopamine content and of secretory responsiveness to hypoxia and hypoglycemia.Secretory responses to strong stimuli, such as high K + or 20% CO 2 , are not significantly affected.We have also shown that, as in the case for other genes relevant to CB function, 14 Olfr78 mRNA expression is strongly downregulated in HIF2α-deficient CB cells. 15Therefore, although Olfr78 is not essential for acute O 2 sensing, it appears to play a role in glomus cell homeostasis.Notably, the characteristics of Olfr78deficient glomus cells resemble those of CB neuroblasts; these are immature glomus cell precursors that complete their maturation upon exposure to hypoxia. 17,40Carotid body neuroblasts are weakly TH-positive and have reduced dopamine quantal content and poor secretory activity and responsiveness to hypoxia but maintain responses to other stimuli.In addition, their capacitance (proportional to cell size) is smaller than that of glomus cells and more than 80% of these cells exhibit large Na + currents, a sign of cellular immaturity. 17,41,42The similarities between Olfr78-deficient cells and CB neuroblasts suggest that Olfr78 exerts a "trophic" role that is required for the maturation of glomus cell phenotype; in this way, a deficiency of Olfr78 causes glomus cells to be in a more immature state.The Olfr78 deficiency may well be compensated for in vivo by other blood borne molecules acting on glomus cells in situ.However, CB glomus cells in in vitro preparations, subjected to enzymatic and/or mechanical stress and maintained for 24-48 h under culture conditions, may have a greater vulnerability to the lack of Olfr78-dependent trophic action.In this sense, it is also logical that the response to hypoxia-a complex process that depends on several steps, involving mitochondrial O 2 sensing and signaling to membrane ion channels-in Olfr78-deficient cells is more affected than the secretion induced by direct depolarization with high extracellular K + .Moreover, the activation of glomus cells with strong stimuli (eg, 40 mm extracellular K + ) favors compound exocytosis, thereby decreasing the number of small events (those more affected in Olfr78-deficient cells) and minimizing the effect of decreased quantal dopamine content on the total secretory activity.
The trophic action of Olfr8 may be related to the need of CB glomus cells to maintain high intracellular levels of cAMP.In addition to some well-known targets of cAMP-dependent transcriptional (CREB-mediated) and post-transcriptional (PKAmediated) signaling pathways essential for glomus cell function, such as Th induction or Ca 2+ channel phosphorylation, respectively, 43,44 there may be many other targets that are yet to be elucidated.Indeed, genes encoding components of cAMP-dependent signaling pathways are among the most highly expressed in glomus cells. 2,3Furthermore, adenylate cyclase activation has a stimulatory action on glomus cells and potentiates the effect of hypoxia. 45Mice with glomus cells deficient in adenylate cyclase 3, a key component of the olfactory signal transduction pathway, exhibit impaired cellular and breathing responses to hypoxia. 46The physiological ligand(s) of Olfr78 is (are) unknown and the possibility that this G-protein-coupled receptor is constitutively activated 47 cannot be discounted.
The contribution of Olfr78 to CB homeostasis suggests that its pharmacological modulation could be of therapeutic and translational interest.Exacerbation of the peripheral chemoreflex and sympathetic overactivation are involved in the pathogenesis of highly prevalent medical conditions, such as refractory hypertension, intermittent hypoxia during sleep apnea, and left cardiac failure. 30Remarkably, sympathetic activation and systemic hypertension induced by chronic intermittent hypoxia are reduced in constitutive Olfr78 knockout mice. 13As that chronic intermittent hypoxia activates the conversion of CB neuroblasts into mature glomus cells, 40 an inhibition of this process could explain the decrease in sympathetic overactivation induced by intermittent hypoxia in constitutive (or global) Olfr78 knockout mice. 13In closing, we speculate that Olfr78 inhibitors, with a tissue or organ action that is more restricted to the CB than HIF2α inhibitors, 48,49 may produce an attenuation of CB activity without entirely abolishing the life-saving responses to hypoxia and hypoglycemia.

A
TH-Olfr78 KO mouse model

Figure 4 .
Figure 4. Passive electrical parameters and voltage-dependent K + currents of dispersed glomus cells of wild-type (WT) mice and TH-Olfr78 KO mice.(A) and (B) Average cell capacitance (in pF, A) and input resistance (in GOhm, B) measured in glomus cells of WT mice and TH-Olfr78 KO mice.Data are presented as boxplots indicating median (middle line), 25th, 75th percentile (box), and largest and smallest values range (whiskers), and with all data values superimposed.Mean ± SEM are: capacitance, WT, 3.955 ± 0.245 (n = 61); KO, 3.589 ± 0.0.251(n = 59); input resistance, WT, 3.298 ± 0.259 (n = 48); KO, 3.726 ± 0.293 (n = 44).(C) Top, representation of the depolarizing pulse protocol (50-ms depolarizing pulses reaching membrane potentials between −30 to +50 mV in steps of 20 mV from a holding potential of −70 mV).Families of representative macroscopic K + currents recorded from dispersed glomus cells of WT mice (middle) and TH-Olfr78 KO mice (bottom) illustrating the variability in the time course of the current in both types of mice.(D) Average maximal K + current density (in pA/pF, ordinate) versus voltage (in mV, abscissa) relationship in glomus cells of WT mice and TH-Olfr78 KO mice.Each point is the average of 11 (WT) and 10 (KO) different experiments.(E) Average maximal K + current density (in pA/pF) recorded in response to depolarizing pulses to +40 mV in isolated glomus cells of WT mice and TH-Olfr78 KO mice.Data are expressed as mean ± SEM.Values are TH+ (WT, 344 ± 64, n = 11; KO, 487 ± 67, n = 10).

Figure 5 .
Figure 5. Voltage-dependent Na + and Ca 2+ currents of dispersed glomus cells of wild-type (WT) mice and TH-Olfr78 KO mice.(A) Macroscopic Na + and Ca 2+ currents elicited by membrane depolarization in a patch clamped glomus cell in the presence (left) and absence (right) of extracellular Na + .Complete replacement of Na + was achieved with the impermeant cation N-methyl glucamine.Na + current (INa) and Ca 2+ current (ICa) were measured at the indicated time.(B) and (C) Representative examples of macroscopic inward currents with variable components of Na + and Ca 2+ current recorded from glomus cells of WT mice (B) and TH-Olfr78 KO mice (C).Standard (full Na + ) external solution and pulse protocol are as illustrated in panel A (left).(D) and (F) Average Na + (D) and Ca 2+ (F) current density (in pA/pF, ordinate) versus voltage (in mV, abscissa) relationship in glomus cells of WT mice and TH-Olfr78 KO mice.Each point is the average of 18 (WT) and 21 (KO) different experiments.(E) and (G) Average Na + (E) and Ca 2+ (G) current density (in pA/pF) recorded in response to depolarizing pulses to +10 mV from glomus cells of WT mice and TH-Olfr78 KO mice.Values, expressed as mean ± SEM, are Na + current (WT, −16.5 ± 2.8, n = 18; KO, −25.8 ± 3.6, n = 21); Ca 2+ current (WT, −14.6 ± 2.9, n = 19; KO, −13.3 ± 2.2, n = 21).P-values (<0.05) calculated with unpaired two-tailed t-test are indicated in panels (E) and (G).
Expression of genes relevant to carotid body (CB) function in TH-Olfr78 KO and Epas1 KO mice.(A) Scheme of the generation of the TH-Olfr78 KO mouse model.(B) and (C) Levels of Olfr78 mRNA, relative to wild type (WT), in CB samples (B) and in superior cervical ganglion (SCG) samples (C) of WT mice and TH-Olfr78 KO mice.Data are expressed as mean ± SEM with individual values superimposed.Values are CB (WT: 1 ± 0.10; KO: 0.00046 ± 0.00008, n = 3 replicates/group); SCG (WT: 1 ± 0.14; KO: 0.0009 ± 0.0001, n = 4 replicates/group).Statistically significant P-values (<0.05) calculated by unpaired two-tailed t-test are indicated.(D) and (E) Levels of mRNA, relative to WT, of genes relevant to glomus cell function: Olfr78, Epas1, Cox8b, Higd1C, and Rgs5 (D) and Cox4i2, Pcx, and Th (E) in CB samples of WT mice and TH-Olfr78 KO mice.Data with normal distribution are presented by bar diagrams (D) and expressed as mean ± SEM with data values superimposed.Values that do not follow a normal distribution are represented as boxplots (E) indicating median (middle line), 25th, 75th percentile (box), and largest and smallest values (whiskers).All data are plotted individually.P-values (<0.05 or near this value) calculated with unpaired two-tailed t-test (D) and with Mann-Whitney test (E) are indicated.
with 5% CO 2 and 95% N 2 to reach an O 2 tension of ∼10 to 15 mmHg in the chamber.Osmolality of solutions was ∼300 mosm/Kg with pH 7.4.The hypercapnic solution was bubbled with a gas mixture of 20% O 2 , 20% CO 2 .The low glucose experiments were done upon exposure of the cells to solutions in which sucrose replaced glucose (125 mM NaCl, 4.5 mM KCl, 23 mM NaHCO 3 , 1 mM MgCl 2 , 2.5 mM CaCl 2 , 0 mM glucose, and 10 mM sucrose or 0.5 mM glucose and 9.5 mM sucrose).